1+ {
2+ "nbformat" : 4 ,
3+ "nbformat_minor" : 0 ,
4+ "metadata" : {
5+ "colab" : {
6+ "name" : " Realtime Text Extracion.ipynb"
7+ "provenance" : [],
8+ "collapsed_sections" : []
9+ },
10+ "kernelspec" : {
11+ "display_name" : " Python 3"
12+ "name" : " python3"
13+ }
14+ },
15+ "cells" : [
16+ {
17+ "cell_type" : " markdown"
18+ "metadata" : {
19+ "id" : " lv8kfltaKySW"
20+ },
21+ "source" : [
22+ " ##**TEXT EXTRACTION FROM A REAL TIME CAPTURED IMAGE**"
23+ ]
24+ },
25+ {
26+ "cell_type" : " markdown"
27+ "metadata" : {
28+ "id" : " wWB6a7k7IQ4W"
29+ },
30+ "source" : [
31+ " ###**Code for starting the webcam in the google colab**"
32+ ]
33+ },
34+ {
35+ "cell_type" : " code"
36+ "metadata" : {
37+ "id" : " SucxddsPhOmj"
38+ },
39+ "source" : [
40+ " from IPython.display import display, Javascript\n "
41+ " from google.colab.output import eval_js\n "
42+ " from base64 import b64decode\n "
43+ " \n "
44+ " def take_photo(filename='photo.jpg', quality=0.8):\n "
45+ " js = Javascript('''\n "
46+ " async function takePhoto(quality) {\n "
47+ " const div = document.createElement('div');\n "
48+ " const capture = document.createElement('button');\n "
49+ " capture.textContent = 'Capture';\n "
50+ " div.appendChild(capture);\n "
51+ " \n "
52+ " const video = document.createElement('video');\n "
53+ " video.style.display = 'block';\n "
54+ " const stream = await navigator.mediaDevices.getUserMedia({video: true});\n "
55+ " \n "
56+ " document.body.appendChild(div);\n "
57+ " div.appendChild(video);\n "
58+ " video.srcObject = stream;\n "
59+ " await video.play();\n "
60+ " \n "
61+ " // Resize the output to fit the video element.\n "
62+ " google.colab.output.setIframeHeight(document.documentElement.scrollHeight,true);\n "
63+ " \n "
64+ " // Wait for Capture to be clicked.\n "
65+ " await new Promise((resolve) => capture.onclick = resolve);\n "
66+ " \n "
67+ " const canvas = document.createElement('canvas');\n "
68+ " canvas.width = video.videoWidth;\n "
69+ " canvas.height = video.videoHeight;\n "
70+ " canvas.getContext('2d').drawImage(video, 0, 0);\n "
71+ " stream.getVideoTracks()[0].stop();\n "
72+ " div.remove();\n "
73+ " return canvas.toDataURL('image/jpeg', quality);\n "
74+ " }\n "
75+ " ''')\n "
76+ " display(js)\n "
77+ " data = eval_js('takePhoto({})'.format(quality))\n "
78+ " binary = b64decode(data.split(',')[1])\n "
79+ " with open(filename, 'wb') as f:\n "
80+ " f.write(binary)\n "
81+ " return filename"
82+ ],
83+ "execution_count" : 4 ,
84+ "outputs" : []
85+ },
86+ {
87+ "cell_type" : " markdown"
88+ "metadata" : {
89+ "id" : " KxGGlrg-K_Wz"
90+ },
91+ "source" : [
92+ " ####Click on the **Capture** button for capturing the image"
93+ ]
94+ },
95+ {
96+ "cell_type" : " code"
97+ "metadata" : {
98+ "id" : " buJCl90WhNfq"
99+ },
100+ "source" : [
101+ " from IPython.display import Image\n "
102+ " try:\n "
103+ " filename = take_photo()\n "
104+ " print('Saved to {}'.format(filename))\n "
105+ " \n "
106+ " # Show the image which was just taken.\n "
107+ " display(Image(filename))\n "
108+ " except Exception as err:\n "
109+ " # Errors will be thrown if the user does not have a webcam or if they do not\n "
110+ " # grant the page permission to access it.\n "
111+ " print(str(err))"
112+ ],
113+ "execution_count" : null ,
114+ "outputs" : []
115+ },
116+ {
117+ "cell_type" : " markdown"
118+ "metadata" : {
119+ "id" : " V-95YMbaIjCL"
120+ },
121+ "source" : [
122+ " ###**Installing and importing the required libraries**"
123+ ]
124+ },
125+ {
126+ "cell_type" : " code"
127+ "metadata" : {
128+ "id" : " TPetMlwdvspP"
129+ },
130+ "source" : [
131+ " !sudo apt install tesseract-ocr\r\n "
132+ " !pip install pytesseract\r\n "
133+ " import pytesseract as ts\r\n "
134+ " import cv2\r\n "
135+ " from google.colab.patches import cv2_imshow"
136+ ],
137+ "execution_count" : null ,
138+ "outputs" : []
139+ },
140+ {
141+ "cell_type" : " code"
142+ "metadata" : {
143+ "id" : " c5UJmjW8wbZ9"
144+ },
145+ "source" : [
146+ " '''\r\n "
147+ " If there is error occured as path is not defined or you get the wrong image\r\n "
148+ " Follow the given steps\r\n "
149+ " Put the path of the picture captured by you\r\n "
150+ " You can find it in the files part under the navigation in the left side\r\n "
151+ " Right click on the photo, click on \" copy path\" and paste it here \r\n "
152+ " '''\r\n "
153+ " \r\n "
154+ " img=cv2.imread('/content/photo.jpg')\r\n "
155+ " gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)\r\n "
156+ " cv2_imshow(img)\r\n "
157+ " #Removing noise from the image\r\n "
158+ " ret, thresh = cv2.threshold(gray,1010, 200, cv2.THRESH_OTSU, cv2.THRESH_BINARY)\r\n "
159+ " text = ts.image_to_string(thresh) \r\n "
160+ " \r\n "
161+ " cv2_imshow(thresh) #printing realtime text\r\n "
162+ " print(\"\\ n\\ n\" )\r\n "
163+ " print(text)\r\n "
164+ " \r\n "
165+ " #If the desried output is not obtained just change the 'thresh to 'gray' in line 13 and 15"
166+ ],
167+ "execution_count" : null ,
168+ "outputs" : []
169+ },
170+ {
171+ "cell_type" : " code"
172+ "metadata" : {
173+ "id" : " fdVdr4Ka05FG"
174+ },
175+ "source" : [
176+ " "
177+ ],
178+ "execution_count" : null ,
179+ "outputs" : []
180+ },
181+ {
182+ "cell_type" : " code"
183+ "metadata" : {
184+ "id" : " jUyVjAV9w3iG"
185+ },
186+ "source" : [
187+ " "
188+ ],
189+ "execution_count" : null ,
190+ "outputs" : []
191+ }
192+ ]
193+ }
0 commit comments